There for the Reaping: The Ethics of Harvesting Online Data for Research Purposes

Online social environments offer a rich source of data that researchers can harvest to gain insight into a wide range of social issues. This type of research is sometimes considered as observation of public behaviour, and therefore exempt from ethical review. This type of research, however, raises ethical issues with respect to the public/private nature of online spaces, consent, and anonymity in the online environment. This project examines research ethics guidelines for recommendations regarding the use of harvested online data, identifying best practices for researchers who engage in this type of research. Introduction Online social environments, including social media platforms and online discussion groups, provide important platforms for individual expression and interpersonal connection. These platforms also provide researchers with a source of data that can be mined for valuable insight into social issues. The advantages of these data for research are myriad, and include limited costs (time and money) for data collection, access to information on sensitive issues, and an absence of reactivity. At the same time, use of these data, which are generated by and about individuals, warrants careful ethical consideration. Research ethics guidelines, including the Tri-Council Policy Statement (TCPS2) in Canada, provide guidance to researchers and research ethics boards (REBs) tasked with making decisions and recommendations regarding the ethical conduct of research involving human participants. This guidance, however, is variable and often incomplete, with the result that ethical considerations and practices vary substantially between researchers and across institutions. This project examines research ethics guidelines, including the TCPS2 and others, with respect to their coverage of and recommendations concerning research involving the harvesting of social media data. The results summarize the treatment of these issues in guidelines, available in English, that apply to research and researchers in Canada and internationally, culminating with a set of best ethical practices for research that involves this type of data. Research Ethics Issues and Guidelines Online social media posts, including tweets, online discussion group interactions, Facebook updates, and listerv archives, provide a rich source of ‘naturally occurring, everyday talk’ (Sixsmith and Murray, 2001, p. 424) – precisely the kind of data that, according to Potter and Wetherell (1995), offer deep insight into social phenomena. Ethical issues arise, however, with any use of data from or about people, and harvested data from online social interactions is no exception. Researchers and those focused on the ethics of human subjects research have long recognized that online research presents new and specific ethical issues, potentially requiring new ethical approaches (see, e.g., Elgesem, 2002; Eysenbach & Till, 2001; Flicker et al., 2004; Neuhaus & Webmoor, 2012; King, 1996; Swirsky et al., 2014; Taylor & Pagliari, 2018, Vitak et al., 2016). The harvesting of online productions and interactions for research purposes presents particular challenges, revolving around the issues of participant autonomy (including questions of consent), and participant anonymity and confidentiality. Under many research ethics guidelines harvesting of online social media data is a form of observational research that, if carried out in ‘public’ venues, may be exempt from ethical review. Questions arise, however, regarding the public/private nature of online spaces – and, if the spaces are deemed ‘private’ (or at least not so ‘public’ as to remove the requirement for ethical review), issues of consent and anonymity/confidentiality become paramount. Some research ethics guidelines (e.g., the TCPS2) provide limited specific guidance regarding this type of online research; others, such as the guidelines provided by the Association of Internet Researchers, (Franzke et al., 2020) provide much more focused and detailed recommendations. Researchers and REBs considering the ethics of this type of research would benefit from a comprehensive and integrated summary of the ethical considerations and approaches available in various guidelines: it is exactly such a summary that we present here. Methodology A total of 31 research ethics guidelines, written in English, were identified through a combination of methods including Google searches (e.g., for ‘research ethics’ and ‘ethics guidelines), review of research using harvested online data for mention of research guidelines, and review of publications addressing online research methods for cited guidelines, and specific search of the websites of North American universities for ethics guidelines.